# Can I do a Multiple Regression with Excel?

I have performance data for 3 years for about 50K employees. Each employee has 3 ratings associated with them. I would like to create an equation that will predict (or try to predict) the future value of that employee's rating given the historical ratings I have available.

My problem is that I am not terribly familiar with complex regression analysis and I'm not quite sure how to set this up in Excel.

The data are laid out as:

`EE ID, Rating 1, Rating 2, Rating 3`

The problem (as I see it) is that Excel expects a Y value as input and I technically don't have any valid Y values. My Y value equivalent would be the `EE ID`, which does not have any relationship to the X values.

Any idea how I can do this?

-
So if EEID1 got ratings of 100, 200, and 300 you'd want it to tell you the next most likely number by calculating the average slope of the increases? Your Ys would be the time periods, yearly – Raystafarian Feb 14 '12 at 17:08

## 1 Answer

You could use `=forecast` by going `=FORECAST(Y4,R1:R3,Y1:Y3)` this is, you would assume that each of the ratings (R1:R3) are associated with each year (Y1:Y3) and you want the next Rating for the next Year (Y4)
you could do nearly the same thing for `Linest, Trend, Slope, Steyx, Longest, and Growth`.

## EDIT

If this was your data

``````Year   2000 2001 2002
Rating 100  100  150
``````

For each EEID and you wanted 2003's forecasted rating, your formula would be `=forecast(2003,B2:B4,A2:A4)`

-
Could you please elaborate a bit more on this answer? Would my setup be something like EE ID | 2012 | Rating 1 | Rating 2 | Rating 3 | 2009 | 2010 | 2011? – Jay Feb 15 '12 at 19:43
@Jay added some explanation – Raystafarian Feb 15 '12 at 19:49
Thank you - I will try this! – Jay Feb 15 '12 at 19:51